We propose to advance time-of-flight (ToF) positron emission tomography (PET) detector instrumentation that, if successful, will further enhance abilities to visualize and quantify molecular signatures of disease in the clinic. ToF PET uses the arrival time difference of detected coincidence photons to better estimate the position of the positron annihilation along the response line between any two detector elements in the PET system. Accurate ToF event positioning requires sub- nanosecond coincidence time resolution to reduce the uncertainty in annihilation photon emission location along a response line. For state-of-the-art clinical ToF PET systems, which achieve ~600-900 ps full-width-at-half-maximum (FWHM) coincidence time resolution [using photomultiplier tubes (PMTs)], the photon depth of interaction (DoI) uncertainty within the e2 cm length detector crystals does not significantly affect ToF position uncertainty. For the proposed d300 ps coincidence time resolution, the ToF uncertainty due to photon DoI within e2 cm length crystals cannot be ignored. Thus, our goal in this proposal is to create a PET detector with d300 ps FWHM coincidence time resolution that also measures photon DoI within the scintillation crystal. In addition to enhancing photon arrival time information, the capability for photon DoI resolution also promotes spatial resolution uniformity across the field of view (FoV). Furthermore, the proposed design has the unique capability to measure the 3D position and energy of each individual interaction of multi-interaction photon events, which can be exploited to further improve spatial resolution and contrast resolution. To achieve these design goals, we propose to explore a new detector design based on single ended readout of e2 cm length scintillation crystals coupled one-to-one to arrays of fast, high-gain silicon photomultiplier (SiPM) photodetectors. The full detector signal waveforms will be digitized by novel, commercially available sampling architectures, and DoI (and 3D positioning) information is determined by correlation with various parameters of the digitized detector pulse shape for each event, such as pulse height, rise and falling edge frequency patterns. In a PET system, DoI information leads to more accurate ToF event positioning along a response line that can impact reconstructed image performance, but in this work we focus on studying dependence of photon arrival time and coincidence time resolution on photon DoI. If the proposed design does not meet the time and DoI resolution specifications, as a backup plan, an alternative detector architecture based on layers of short scintillation detectors will be studied. Impact: If successful, a PET system built with the proposed detectors will increase image signal-to-noise ratio (SNR) three-fold compared to a non-ToF system for a 40 cm diameter patient, and provide enhancement of spatial resolution and contrast resolution that together will substantially enhance the ability to visualiz and quantify molecular signatures of disease residing in diffuse background activity. Alternatively the substantial SNR boost can be exploited to reduce injected dose or scan time. PUBLIC HEALTH RELEVANCE: We propose to develop an advanced, yet practical photon detector technology appropriate for a new-generation clinical time-of-flight positron emission tomography (PET) system that has better than 300 pico-seconds coincidence (two-photon) time resolution and 5 mm photon interaction depth resolution within the entire detector volume. If successful, such an advance would enable substantial enhancements to image quality and quantitative accuracy over current PET system technology that would translate into benefits such as (1) improved visualization and quantification of subtle molecular and cellular-based signatures of disease residing in a diffuse background, or (2) substantially reduced injected radiation dose and/or scan duration. These features would both help to promote more widespread use of PET as well as expand its role in the clinical management of disease.